On 20/01/14 23:29, Joseph Holsten wrote:
> Anyone know what would be involved in reviving the patch from long ago adding
> disk io to the core metrics?[1]
>
> I see that there's something similar for solaris[1], but digging around in
> the solaris impl[2] shows this is coming from a kstat_read(3KSTAT), so that
> isn't going to work for linux.
>
> Apparently this patch grabs the data from /proc/partitions or /proc/diskstats
> depending on kernel version. This seems to be supported by the linux kernel
> iostats docs[3]. Does that seem sane?
>
> I've never written a C plugin for ganglia before, so if you've got any
> advice, I'd appreciate it.
> --
Joseph, please have a look at the IO module in ganglia-modules-linux:
https://github.com/ganglia/ganglia-modules-linux
It is based on a patch JB Kim contributed for Ganglia 3.0.x, I converted
it to a module
It currently does aggregate IO stats from all disks, it could be
enhanced to work on a per-disk basis
------------------------------------------------------------------------------
WatchGuard Dimension instantly turns raw network data into actionable
security intelligence. It gives you real-time visual feedback on key
security issues and trends. Skip the complicated setup - simply import
a virtual appliance and go from zero to informed in seconds.
http://pubads.g.doubleclick.net/gampad/clk?id=123612991&iu=/4140/ostg.clktrk
_______________________________________________
Ganglia-developers mailing list
Ganglia-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-developers
